Research Article Critical factors of indirect determination of 3-chloropropane- 1,2-diol esters

Transcription

1 Eur. J. Lipid Sci. Technol. 2011, 000, esearch Article Critical factors of indirect determination of 3chloropropane 1,2diol esters Karel Hrncirik 1, Zuzana Zelinkova 2 and Alessia Ermacora 1 1 Unilever &D Vlaardingen, Vlaardingen, The Netherlands 2 Institute of Chemical Technology, Prague, Czech epublic Several indirect methods have been developed for the determination of 3chloropropane1,2diol (3MCPD) esters in fat matrices, but the variation between these methods leads to an uncertainty about the comparison of the results. All the indirect methods are based on the conversion of 3MCPD esters into free 3MCPD upon transesterification (either in acid or alkaline solution), its purification, derivatisation and instrumental analysis. In this study, major critical factors of the methodology, particularly transesterification, salting out and the choice of internal standard, were evaluated. 3MCPD showed low stability under the conditions of alkaline transesterification (only around 40% recovery) affecting the sensitivity of the method. Furthermore, the results obtained within the first 1 2 min were 10 20% higher than those obtained at longer (5 10 min) transesterification time. A comparison of results obtained by using two different internal standards (free or esterified form of deuterated 3MCPD) revealed a slight overestimation (7 15%) when free deuterated 3MCPD was used, which may be possibly explained by the different behaviour of the both forms during alkaline transesterification. The use of different salts and ph values of the reaction medium during salting out step had a major impact on the results due to the formation of additional 3MCPD. ther compounds, like glycidyl esters, present in oil samples may be converted to 3MCPD during this step, however, the conversion was not complete and varied substantially (12 93%) in dependence on the ph value of the medium. The method based on acid transesterification showed better selectivity and robustness. Keywords: Analysis / Chloropropanediols / Edible oils / 3MCPD ester / Transesterification eceived: September 17, 2010 / evised: December 7, 2010 / Accepted: December 21, 2010 DI: /ejlt Introduction 3Chloropropane1,2diol (3MCPD) esters represent a class of foodborne contaminants that are mainly formed during hightemperature processing of fatbased matrices. Traces of 3MCPD esters were first detected by Velíšek et al. in 1980 [1] in acidhydrolysed vegetable proteins together with other chloropropanols, but they were not of Correspondence: Dr. Karel Hrncirik, Unilever &D Vlaardingen, Department of Fat Technology, livier van Noortlaan AT Vlaardingen, The Netherlands karel.hrncirik@unilever.com Fax: þ Abbreviations: 3MCPD, 3chloropropane1,2diol; 3MCPDd5, deuterated 3chloropropane1,2diol; PP3MCPD, 1,2dipalmitoyl3chloropropane; PP3MCPDd5, deuterated 1,2dipalmitoyl3chloropropane; PBA, phenylboronic acid general concern till 2004, when their occurrence was reported in processed food [2]. esulting from the review of toxicology, mutagenicity and carcinogenicity data in short and longterm toxicity studies by the Joint FA/WH Expert Committee on Food Additives, a provisional maximum tolerable daily intake of 2 mg 3MCPD/ kg body weight was assigned [3]. The level of free 3MCPD in soy sauce and hydrolysed vegetable protein has been regulated by setting up a limit in these products [4], assuming these are the main contributors to the dietary intake. In vitro studies have demonstrated that 3MCPD mono and diesters of fatty acids are hydrolysed by the intestinal lipase to release 3 MCPD [5]. Due to the lack of in vivo data the risk assessment of 3MCPD esters is currently carried out under the assumption that 3MCPD is completely cleaved from its esters. Despite the development of a few analytical methods for the determination of free 3MCPD in various matrices, no standard method has been established. The authorities only defined performance criteria that a method must fulfil in order to provide officially acceptable results [6].

2 2 K. Hrncirik et al. Eur. J. Lipid Sci. Technol. 2011, 000, Within last few years several approaches have been reported for the analysis of 3MCPD esters (also referred as bound 3 MCPD ). They are mostly based on indirect methods that involve the release of 3MCPD from its esters by transesterification in acid [7] or alkaline medium [8], or enzymatically [9], followed by the purification, derivatisation and quantification of the free form. Modifications of each individual step by different laboratories lead to a number of inhouse methods and, consequently, to a wide variation in results (Karasek, T. Wenzl, F. Ulberth: Proficiency test on the determination of 3MCPD esters in edible oil. JC Scientific and Technical eports, 2010, EU EN. html/interlaboratory_comparisons/3_mcpd/index.htm). There is an urgent need for a reliable validated method, which would fulfil the basic requirements, namely sensitivity, specificity, trueness and robustness. The aim of this study was to identify and evaluate the critical factors affecting the indirect determination of 3 MCPD esters, and particularly to compare the two basic analytical approaches (i.e. alkaline and acid transesterification) with respect to their impact on the results. 2 Materials and methods 2.1 eagents and chemicals 1,2Dipalmitoyl3chloropropane (PP3MCPD, purity 99.8%) and deuterated 1,2dipalmitoyl3chloropropane (PP3 MCPDd5, purity 99.0%) were synthesised according to Kraft et al. [10] and purified on silica gel column. 3 MCPD (purity 98%), sodium sulphate (purity 99.0%), ammonium sulphate (purity 99.5%) and sodium chloride (purity 99.5%) were purchased from Merck (Darmstadt, Germany), deuterated 3chloropropane1,2diol (3MCPDd5, purity 98%) from CDN Isotopes (PointeClaire, QC, Canada), glycidol (purity 96%) from Sigma Aldrich (Bellefonte, PA, USA), phenylboronic acid (PBA, purity 97%), sodium methoxide (purity 97%) and sulphuric acid (95%) from Fluka (Buchs, Switzerland) and sodium hydrogencarbonate (purity 99.5%) from VW Intl. (West Chester, PA, USA). All solvents used were of analytical grade. 2.2 Samples Physically refined palm oil samples (containing significant levels of 3MCPD esters), as well as crude (degummed) rapeseed oil (3MCPD esters not detected) were obtained from local suppliers. Crude rapeseed oil was spiked with PP3MCPD (4.98 mg/kg, expressed as bound 3MCPD) or with glycidol ( mg/kg). 2.3 Methods Indirect analysis of 3MCPD esters in oils was performed using two different procedures. Previously developed methods [11, 12] were adopted and modified as described in Sections and 2.3.2, respectively Method A (involving alkaline transesterification) ne hundred milligram (5 mg) of oil was weighed in a glass tube and dissolved in 500 ml of tertbutylmethylether. Twenty microlitre of internal standard solution (3MCPDd5 or PP3MCPDd5) was added. The cleavage of 3MCPD esters was achieved by addition of 200 ml of sodium methoxide in methanol (0.5 mol/l). The mixture was allowed to stand at room temperature for min. The reaction was stopped by addition of 0.6 ml of acidic ammonium sulphate solution (37.7% w/v ammonium sulphate aqueous/25% sulphuric acid, 50:3 v/v). Besides ammonium sulphate, sodium chloride was also tested as salting out agent. FAME were extracted from the sample by repeated liquid liquid extraction with isohexane (2 1 ml). Free 3MCPD was extracted by subsequent liquid liquid extraction with ethylacetate (2 0.6 ml) and derivatised by 100 ml of PBA solution (saturated, in diethylether); the reaction was facilitated by sonication for 2 3 min. The sample was evaporated to dryness under a nitrogen stream and the residue was dissolved in ml of acetone prior to GCMS analysis Method B (involving acid transesterification) ne hundred milligram (5 mg) of oil was weighed in a glass tube and dissolved in 1 ml of tetrahydrofuran containing the internal standard (3MCPDd5 or PP3MCPDd5). 1.8 ml of sulphuric acid solution in methanol (1.8% v/v) was added to the sample and the mixture was incubated at 408C for 16 h. The reaction was stopped by addition of 0.5 ml sodium hydrogencarbonate solution (saturated, aqueous) and the organic solvents were evaporated under a nitrogen stream. FAME were separated from the sample by addition of 2 ml of aqueous salt solution (20% w/v; sodium sulphate was used in the standard procedure, sodium chloride was tested as an alternative) followed by liquid liquid extraction with hexane (2 2 ml). Two hundred and fifty microlitre of PBA solution (25% w/v, acetone/water, 19:1 v/v) was added to the reaction mixture, which was incubated at 808C for 20 min. The 3MCPDPBA derivative was extracted by 1 ml hexane and analysed by GCMS GCMS analysis GCMS analysis of 3MCPD derivatised by PBA was carried out on an Agilent Technologies 7890A gas chromatograph (Agilent Technologies, Palo Alto, CA, USA) equipped with a quadrupole mass selective detector Agilent 5975C MSD (70 ev) and data processing system MSD Productivity ChemStation. GC was performed on a bonded, poly (dimethylsiloxane) capillary column (Supelco Equity1,

3 Eur. J. Lipid Sci. Technol. 2011, 000, Critical factors of indirect determination of 3chloropropane1,2diol esters 3 30 m length 0.25 mm id 1.0 mm film thickness). The injector temperature was set to 2508C. The GC oven temperature was initially kept at 808C for1minandthenraisedto 1708C at108c/min gradient, then to 2008Cat38C/min, then to 2808C at158c/min and held for 15 min at the final temperature. Helium (flow rate 0.8 ml/min) was used as carrier gas. ne microlitre of sample was injected (pulsed splitless mode). For quantification purposes, ions at m/z 147 (3MCPD) and m/z 150 (3MCPDd5) were chosen for single ion monitoring. Ions at m/z 196, 198 (3MCPD) and m/z 201, 203 (3MCPDd5) were used as qualifiers. 3 esults and discussion The basic principle of indirect determination of bound 3 MCPD is the conversion of a number of individual 3MCPD esters of fatty acids into a single compound, 3MCPD, that is quantified. Currently a large number of analytical methods used by different laboratories exist. The analytical protocol of all these methods comprises a uniform sequence of steps: addition of an internal standard to the sample (either free or esterified form of 3MCPDd5), transesterification (performed either in acid or alkaline medium at different reaction time), followed by neutralisation of the reaction mixture and salting out (using different neutralising reagents and salts), derivatisation of the extracted 3MCPD (mostly by PBA or heptafluorobutyrylimidazol) and GC MS analysis. There are variations within each individual analytical step that may have a significant impact on the specificity, repeatability, reproducibility, trueness and other parameters of the method. A recent proficiency test demonstrated an impact of these variations on the results (Karasek, T. Wenzl, F. Ulberth: Proficiency test on the determination of 3MCPD esters in edible oil. JC Scientific and Technical eports, 2010, EU EN. comparisons/3_mcpd/index.htm). Using the inhouse methods for the analysis of naturally contaminated palm oil, only 19 out of 34 participating laboratories (which included commercial, industrial and official control laboratories) provided results which were considered satisfactory. There is a need for closer investigation of the individual steps and their optimisation in order to achieve overall harmonisation of the analytical methodology. 3.1 Decomposition of 3MCPD during transesterification The first step of the analytical protocol in the indirect 3 MCPD determination is the transesterification in the presence of methanol (methanolysis), which results in the conversion of TAGs and partial acylglycerols into FAME and glycerol. Simultaneously, 3MCPD esters are converted to free 3MCPD. The transesterification step has been proposed both in acid [7] and alkaline [8] media. It is has been shown that 3MCPD is unstable in alkaline solutions [13, 14] giving rise to glycerol (Fig. 1). This work was therefore focused on studying the stability of 3MCPD esters under the conditions of alkaline transesterification employing method A (see Section for details). The addition of the internal standard (3MCPDd5) after transesterification of the sample enabled us to quantify the recovery of 3MCPD released from its native esters. An initial experiment performed in a model system indicated that 73% of 3MCPD (released from a standard solution of PP3MCPD) was decomposed within 10 min of transesterification. Next, the recovery was evaluated by using two oil samples crude rapeseed oil spiked by PP3MCPD and refined palm oil. The samples were analysed after 1, 10 and 30 min of transesterification (Table 1). Cl C 1monoester 3MCPD C H 2 Cl 1,2diester 3MCPD C C Cl 3MCPD HCl glycidol glycerol Cl 2monoester 3MCPD Figure 1. Mechanism of 3MCPD esters transformation under conditions of alkaline transesterification

4 4 K. Hrncirik et al. Eur. J. Lipid Sci. Technol. 2011, 000, Table 1. ecovery of 3MCPD found at different periods of alkaline transesterification Method a) Alkaline transesterification (method A) Acid transesterification (method B) 1 min 10 min 30 min 16 h 3MCPD ecovery 3MCPD ecovery 3MCPD ecovery 3MCPD ecovery Spiked rapeseed oil b) n.d. c) efined palm oil d) n.d. c) a) For the details of the methods see sections and b) Sample spiked at the level of 4.98 mg/kg bound 3MCPD. c) Limit of detection: 0.08 mg/kg. d) ecovery calculated from the mean value of the repeated analysis (N > 10) by method B. The results showed a considerable decrease of 3MCPD within the first 10 min of alkaline transesterification. The recoveries were slightly higher, though comparable, to those found in the model system. After 30 min of alkaline transesterification 3MCPD was not detected. The small difference (12%) in recovery levels between both samples found after 1 min of transesterification could be due to the fact that the refined palm oil sample contains a mixture of 3MCPD fatty acid esters, presumably both mono and diesters in a typical ratio (15:85) as reported by Seefelder et al. [5], whereas the spiked rapeseed oil sample contains only PP 3MCPD. It is reasonable to expect that the conversion of 3 MCPD monoesters into 3MCPD proceeds more rapidly than the one of diesters and that 1 min reaction time may not be sufficient to complete the transesterification, which possibly results in the higher recovery of 3MCPD in the sample containing monoesters in the initial phase of the reaction. The analysis of both samples was also carried out using the acid transesterification (method B, Section 2.3.2). In this case the degradation of 3MCPD was not observed. Although an advantage of the alkaline transesterification is its short duration, it appears that the transesterification time has a major impact on the sensitivity of the method due to the degradation of 3MCPD in alkaline medium. It can be expected that the reaction times prescribed by the methods developed, i.e min [15] and 10 min [11], affect the sensitivity of these methods significantly (based on the data obtained, the sensitivity may be reduced approximately twotimes within this range). Thus there may be scope to optimise the method to achieve maximum recovery by minimising the transesterification time. 3.2 Effect of transesterification time and internal standard Assuming that the decomposition of 3MCPD (released from its native esters) and the deuterated analogue (used as internal standard) occurs at the same rate during alkaline transesterification, the transesterification time should not have an impact on the trueness of the analytical method. In order to check the validity of this assumption, a sample of refined palm oil was repeatedly analysed according to the method A using variable time of transesterification (from 0.5 to 10 min). PP3MCPDd5, added at the beginning of the transesterification (as routinely performed), was used as internal standard in order to minimise the difference between the analyte (in refined oils 3MCPD occurs mainly in the form of diesters) and the internal standard. The 3MCPD levels measured within the first 1 2 min were repeatedly 10 20% higher than those obtained at longer (5 10 min) transesterification times (see Fig. 2 as an example). This surprising finding suggests different behaviour of the analyte and the internal standard at the beginning of the reaction. This may be due to different conversion rates of the individual 3MCPD esters (e.g. monoesters being converted faster as proposed in Section 3.1). Nevertheless, at transesterification times longer than 3 5 min the measured level were constant. 3MCPD esters [mg/kg] transesterification time [min] Figure 2. Effect of transesterification time ( min) on the content of 3MCPD esters in refined palm oil analysed by alkaline transesterificationbased method (method A)

5 Eur. J. Lipid Sci. Technol. 2011, 000, Critical factors of indirect determination of 3chloropropane1,2diol esters 5 If it is the case that individual 3MCPD species (free 3 MCPD, mono and diesters of fatty acid) behave differently, it may be expected that the choice of the internal standard will affect the results. Different laboratories use deuterated forms of either free or esterified 3MCPD (with PP3MCPDd5 being probably the most common). Free 3MCPDd5 is used as an internal standard in validated methods [11, 15] most likely because it was the only suitable compound commercially available at the time of the validation work. Both internal standards (3MCPDd5 and PP3MCPDd5) were tested in both methods (A and B) performed under standard conditions (transesterification 10 min and 16 h, respectively). The results showed that the application of 3 MCPDd5 led to higher results (7 15%) of 3MCPD esters measured by method A. This suggests, again, different behaviour (stability and/or transesterification rate) of free and esterified form of 3MCPD during alkaline transesterification. Taking into account the difference found and the fact that in oils and fats 3MCPD is predominantly present in its bound forms (mainly as fatty acid diesters) we incline towards using PP3MCPDd5 rather than the free form, which occurs only in trace amounts in fatbased matrices. 3.3 Effect of salting out Another critical step that requires particular attention is salting out. This step follows the transesterification (performed either in alkaline or acid media) and facilitates the extraction of FAME (produced during the transesterification) from the reaction mixture. The most common salting out agents used in the analysis of 3MCPD are sodium chloride [7, 8] or sulphate salts [11]. These are added to the reaction mixture either after or during the neutralisation. In this study two salts, sodium chloride and ammonium sulphate, were alternately used as salting out agents in both methods, which were carried out under standard conditions. The choice of these salts was determined by the concern that the introduction of a source of chloride (as NaCl) in the transesterified mixture could lead to the formation of 3 MCPD ex novo (for more details see Section 3.4). The 3 MCPD levels found in spiked rapeseed oil and refined palm oil samples are given in Table 2. The results obtained by the method A show dependence on the type of salt, whereas no significant differences were found when the method B was used. This is particularly obvious in the case of refined palm oil. Such an outcome suggests that the reaction mixture of the alkaline transesterification contains compounds that react with chloride ions to give rise to additional 3MCPD, which leads to the overestimation of the results. In order to understand the reactivity of chloride ion with 3MCPD precursors during sample preparation, the ph value of the reaction mixture was checked after transesterification and neutralisation. Contrary to expectations, the methods employing alkaline transesterification had considerable variation in the ph value of neutralised mixture, (method A) and 6 7 [15], while the ph for the acid transesterificationbased method (method B) was around 7. To verify whether the additional formation of 3MCPD from its precursors (a plausible explanation of the result variation reported in Table 2) is affected by different conditions of alkaline transesterification/neutralisation, a sample of refined palm oil was subjected to the analysis by method A, which was modified in such a way that the ph value in the salting out step was adjusted within the range Both salting out agents sodium chloride or ammonium sulphate were used and the levels found were plotted against the ph value (Fig. 3). In accordance with the previous results the data indicate the yield of 3MCPD is higher (in comparison to ammonium sulphate) by approximately 20% at ph above 3 when sodium chloride was used for salting out. Moreover, the conversion of its precursors to 3MCPD seems to be phdependent: the level of 3MCPD found in the palm oil sample almost doubled when the ph value of the reaction mixture was below 3, which indicates considerably higher conversion rate of glycidol into 3MCPD in the acid environment. 3.4 Method specificity In 2008, the specificity of the newly developed method based on alkaline transesterification and salting out by sodium chloride [8] was questioned by demonstrating that the content of 3MCPD esters is overestimated in some samples Table 2. Effect of salting out on the content of 3MCPD esters in refined palm oil Method a) Method A Method B Salting out agent NaCl (NH 4 ) 2 S 4 NaCl (NH 4 ) 2 S 4 3MCPD 3MCPD 3MCPD 3MCPD Spiked rapeseed oil b) efined palm oil a) For the details of the methods see Sections and b) Sample spiked at the level of 4.98 mg/kg bound 3MCPD.

6 6 K. Hrncirik et al. Eur. J. Lipid Sci. Technol. 2011, 000, MCPD esters [mg/kg] ph Figure 3. Effect of ph value and different salts on the content of 3 MCPD esters in refined palm oil analysed by alkaline transesterificationbased method (method A); ^ symbol represents sodium chloride, S symbol represents ammonium sulphate due to the additional 3MCPD formed de novo during the sample preparation [16]. It was suggested [16], and later confirmed [17], that the compound leading to overestimated levels is glycidol (2,3epoxy1propanol), which is present in refined oils as esters of fatty acids. This finding was later utilised in the development of a methodology for the indirect estimation of the level of glycidyl esters in oils. Its principle is the calculation of the difference between two independent determinations: one (X) that is based on the quantification of the sum of 3MCPD and glycidyl esters (after their conversion to 3MCPD) and the other one (Y) that is specific to 3 MCPD esters. Such an approach will provide meaningful results only if the former determination (X) ensures a complete conversion of glycidol into 3MCPD, whilst in the specific method (Y) any such glycidol conversion during the sample preparation is avoided. The latter can be achieved either by replacing sodium chloride by other salt during salting out or by the deliberate decomposition of glycidyl esters prior to transesterification [15]. The conversion of glycidol under the conditions of salting out was studied in rapeseed oil spiked with glycidol at three different levels. Following the transesterification according to the method A, the neutralisation was performed in such a manner that the final ph value of the mixture was either 2 or 7. Sodium chloride was used for salting out. From the obtained analytical results, the conversion of glycidol into 3MCPD was calculated (Table 3). The glycidol conversion in the neutral solution was limited, but was considerably higher at ph 2. This could provide an explanation of the results obtained by the analysis of refined palm oil (which supposedly contains significant level of glycidyl esters) at different ph (Fig. 3). The results also suggest a concentration dependence of the conversion: at low glycidol concentration higher conversion rate was found. In summary, the conditions of the salting out step (particularly ph value) seem to have a significant impact on the conversion of glycidol. The ph value of the neutralised mixture varies between different methods substantially, and although a complete conversion of glycidol is assumed, it is very probable that this will not be always the case and that the results will depend on the method used. Until recently, the indirect differential determination has been the only option to quantify the content of glycidyl esters in oils. It was applied in several works studying the oil refining process, e.g. [17, 18]. In view of the current findings, the results obtained by indirect differential quantification of glycidyl esters should be interpreted carefully and taken as estimation rather than exact levels. It would be desirable to compare the differential approach for glycidyl ester determination with the direct method that has been developed recently [19]. 4 Conclusions Indirect methods for the determination of 3MCPD esters require a sequence of steps, in which the native esters are transformed into the 3MCPD derivative. In this study, we focused on the two initial steps, transesterification and salting out. It was found that the time and other factors (e.g. ph value) of these steps may have a considerable impact on the recovery, specificity and trueness of the method used. The method based on alkaline transesterification is particularly prone to condition variation, whilst transesterification in acid medium avoids the problem with selectivity because of the irreversible degradation of glycidyl esters that cannot be Table 3. Effect of ph value on the conversion of glycidol to 3MCPD in spiked crude rapeseed oil measured by alkaline transesterificationbased method Sample Glycidol level, spiked 3MCPD level /method A a) ph 2 Conversion b) ph 7 Conversion b) Spiked rapeseed oil Spiked rapeseed oil Spiked rapeseed oil a) Transesterification time 10 min.; for further details of the methods see Section b) The factor used for the conversion of glycidol conc. to 3MCPD conc. was 1.49.

7 Eur. J. Lipid Sci. Technol. 2011, 000, Critical factors of indirect determination of 3chloropropane1,2diol esters 7 subsequently converted to 3MCPD (during salting out) and hence interfere with the analysis. Due to the greater robustness shown, the acid transesterification approach seems to be more advantageous that the one based on alkaline transesterification, although this does not necessarily imply a poor repeatability and trueness of the latter one, provided the analysis is carried out strictly under the correct analytical protocol. It is recommendable to use PP3MCPDd5 (rather than 3MCPDd5), as internal standard as it follows the same transesterification/degradation pattern of native 3MCPD diesters. Further review and critical comparison of current methods for the determination of 3MCPD esters would be very beneficial. The progress in the development of reliable methodology will hopefully enable the elucidation of the mechanism of formation of 3MCPD esters in oils and testing different mitigation strategies. The authors have declared no conflict of interest. eferences [1] Velíšek, J., Davídek, J., Kubelka, V., Janícek, G. et al., New chlorinecontaining organic compounds in protein hydrolysates. Agric. Food Chem. 1980, 28, [2] Svejkovská, B., Novotný,., Divinová, V., éblová, Z. et al., Esters of 3chloropropane1,2diol in foodstuffs. Czech. J. Food Sci. 2004, 22, [3] JEFCA, Joint FA/WH Expert Committee on Food Additives: Safety evaluation of certain food additives and contaminants (prepared by 57th Meeting, 5 14 June, 2001, ome, Italy). WH Food Addit. Ser. 2002, 48, [4] Commission egulation (EC) No 1881/2006 of 19 December 2006 setting maximum levels for certain contaminants in foodstuffs (amending Commission egulation (EC) No 466/2001). ff. J. Eur. Union. 2006, L364, [5] Seefelder, W., Varga, N., Studer, A., Williamson, G. et al., Esters of 3chloro1,2propanediol (3MCPD) in vegetable oils: Significance in the formation of 3MCPD. Food Addit. Contam. A 2008, 25, [6] Commission egulation (EC) No 333/2007 of 28 March 2007 laying down the methods of sampling and analysis for the official control of the levels of lead, cadmium, mercury, inorganic tin, 3MCPD and benzo(a)pyrene in foodstuffs. ff. J. Eur. Union. 2007, L88, [7] Divinová, V., Svejkovská, B., Doležal, M., Velíšek, J., Determination of free and bound 3chloropropane1,2diol by gas chromatography with mass spectrometric detection using deuterated 3chloropropane1,2diol as internal standard. Czech. J. Food Sci. 2004, 22, [8] Weißhaar,., Determination of total 3chloropropane1,2 diol (3MCPD) in edible oils by cleavage of MCPD esters with sodium methoxide. Eur. J. Lipid Sci. Technol. 2008, 110, [9] Hamlet, C. G., Sadd, P. A., Chloropropanols and their esters in cereal products. Czech. J. Food Sci. 2004, 22, [10] Kraft,., Brachwitz, H., Etzold, G., Langen, P., Zöpfl, H.J., Massenspektrometrische Strukturuntersuchung stellungsisomerer Fettsaureester der Halogenpropandiole (Desoxyhalogenglyceride). J. Prakt. Chem. 1979, 321, [11] Fry, H., Woehrlin, F., PreissWeigert, A., Collaborative validation study of newly developed gas chromatography mass spectrometry methods for the determination of 3chloro1,2propanediol fatty acid esters in edible oil and fat, Bf Method 9 MCPDFAEsters, version 1, manuscript in preparation. [12] Zelinková, Z., Svejkovská, B., Velíšek, J., Doležal, M., Fatty acid esters of 3chloropropane1,2diol in edible oils. Food Addit. Contam. 2006, 23, [13] Hamlet, C. G., Sadd, P. A., Kinetics of 3chloropropane1,2 diol (3MCPD) degradation in high temperature model systems. Eur. Food es. Technol. 2002, 215, [14] Sim, C. W., Muhammad, K., Yusof, S., Bakar, J., Hashim, D. M., The optimization of conditions for the production of acidhydrolysed winged bean and soybean proteins with reduction of 3monochloropropane1,2diol (3MCPD). Int. J. Food Sci. Technol. 2004, 39, [15] Deutsche Gesellschaft für Fettwissenschaft: DGF Standard Method C III 18 Determination of esterbound 3chloropropane1,2diol (3MCPD esters) and 3MCPD forming substances in fats and oils by means of GCMS, Deutsche Einheitsmethoden zur Untersuchung von Fetten, Fettprodukten,Tensiden und verwandten Stoffen, Wissenschaftliche Verlagsgesellschaft, Stuttgart, Germany [16] Kuhlmann, J., Überbefunde bei der Bestimmung von 3 MCPDEstern in Ölen & Fetten? Mögliche Ursachen und Konsequenzen, ral presentation, 2nd Workshop on Analysis of 3MCPDesters in Edible ils, Federal Institute for isk Assesment (Bf), Berlin, Germany, November [17] Weißhaar,., Perz,., Fatty acid esters of glycidol in refined fats and oils. Eur. J. Lipid Sci. Technol. 2010, 112, [18] Hrncirik, K., van Duijn, G., An initial study on the formation of 3MCPD esters during oil refining. Eur. J. Lipid Sci. Technol. 2011, 113, DI: /ejlt [19] Masukawa, Y., Shiro, H., Nakamura, S., Kondo, N. et al., A new analytical method for the quantification of glycidol fatty acid esters in edible oils. J. leo Sci. 2010, 59,